Unified AI framework to uncover deep interrelationships between gene expression and Alzheimer's disease neuropathologies.
Unified AI framework to uncover deep interrelationships between gene expression and Alzheimer's disease neuropathologies.
复制标题
DOI：
10.1038/s41467-021-25680-7
复制
发表时间：
2021-09-10
影响因子：
16.6
通讯作者：
Lee SI
中科院分区：
文献类型：
--
作者：
Beebe-Wang N;Celik S;Weinberger E;Sturmfels P;De Jager PL;Mostafavi S;Lee SI
深度神经网络（DNN）捕捉变量之间的复杂关系，但是，由于它们需要大量的样本，因此它们的潜力尚未被充分挖掘，以了解基因表达与人类表型之间的关系。在这里，我们介绍了一个分析框架，即MD-AD（用于阿尔茨海默病神经病理学的多任务深度学习），它利用了DNN和多队列设置之间意想不到的协同作用。在这些情况下，使用传统的统计方法可能会阻碍真正的关节分析，传统的统计方法需要“协调”的表型，并且往往会捕获群体水平的变化，从而掩盖更微妙的真实疾病信号。相反，MD-AD结合了在队列中稀疏测量的相关表型，并学习了使用线性模型未发现的基因和表型之间的相互作用，识别出比队列水平变化更微妙的信号，这些信号可以在动物模型和组织中唯一重现。我们发现MD-AD利用了小胶质细胞免疫反应和神经病理学之间的性别特异性关系，为炎症基因和阿尔茨海默病之间的关联提供了细致入微的背景。阿尔茨海默病的分子基础由于大脑基因表达数据的异质性和稀缺性而变得模糊，这限制了复杂模型的有效性。在这里，作者介绍了一个多任务深度学习框架，以学习基因表达和神经病理学之间的可概括和细微差别的关系。
Deep neural networks (DNNs) capture complex relationships among variables, however, because they require copious samples, their potential has yet to be fully tapped for understanding relationships between gene expression and human phenotypes. Here we introduce an analysis framework, namely MD-AD (Multi-task Deep learning for Alzheimer’s Disease neuropathology), which leverages an unexpected synergy between DNNs and multi-cohort settings. In these settings, true joint analysis can be stymied using conventional statistical methods, which require “harmonized” phenotypes and tend to capture cohort-level variations, obscuring subtler true disease signals. Instead, MD-AD incorporates related phenotypes sparsely measured across cohorts, and learns interactions between genes and phenotypes not discovered using linear models, identifying subtler signals than cohort-level variations which can be uniquely recapitulated in animal models and across tissues. We show that MD-AD exploits sex-specific relationships between microglial immune response and neuropathology, providing a nuanced context for the association between inflammatory genes and Alzheimer’s Disease. The molecular basis of Alzheimer’s Disease has been obscured by heterogeneity and scarcity of brain gene expression data, which limit effectiveness in complex models. Here, the authors introduce a multi-task deep learning framework to learn generalizable and nuanced relationships between gene expression and neuropathology.
登录
查看更多内容
影响因子：
8.8
作者：
Mathys H;Adaikkan C;Gao F;Young JZ;Manet E;Hemberg M;De Jager PL;Ransohoff RM;Regev A;Tsai LH
通讯作者：
Tsai LH
影响因子：
9.8
作者：
Allen, Mariet;Carrasquillo, Minerva M.;Funk, Cory;Heavner, Benjamin D.;Zou, Fanggeng;Younkin, Curtis S.;Burgess, Jeremy D.;Chai, High-Seng;Crook, Julia;Eddy, James A.;Li, Hongdong;Logsdon, Ben;Peters, Mette A.;Dang, Kristen K.;Wang, Xue;Serie, Daniel;Wang, Chen;Thuy Nguyen;Lincoln, Sarah;Malphrus, Kimberly;Bisceglio, Gina;Li, Ma;Golde, Todd E.;Mangravite, Lara M.;Asmann, Yan;Price, Nathan D.;Petersen, Ronald C.;Graff-Radford, Neill R.;Dickson, Dennis W.;Younkin, Steven G.;Ertekin-Taner, Nilufer
通讯作者：
Ertekin-Taner, Nilufer
影响因子：
11.2
作者：
Chibnik, Lori B.;Shulman, Joshua M.;Leurgans, Sue E.;Schneider, Julie A.;Wilson, Robert S.;Tran, Dong;Aubin, Cristin;Buchman, Aron S.;Heward, Christopher B.;Myers, Amanda J.;Hardy, John A.;Huentelman, Matthew J.;Corneveaux, Jason J.;Reiman, Eric M.;Evans, Denis A.;Bennett, David A.;De Jager, Philip L.
通讯作者：
De Jager, Philip L.
影响因子：
4.6
作者：
Lee, Taesic;Lee, Hyunju
通讯作者：
Lee, Hyunju
影响因子：
8.8
作者：
Matarin, Mar;Salih, Dervis A.;Edwards, Frances A.
通讯作者：
Edwards, Frances A.